Journal of Minerals & Materials Characterization & Engineering, Vol. 10, No.12, pp.1179-1188, 2011
jmmce.org Printed in the USA. All rights reserved
1179
Taguchi Technique for the Simultaneous Optimization of Tribological
Parameters in Metal Matrix Composite
Hemanth Kumar.T.R.
1
Swamy.R.P.
2
and Chandrashekar T.K.
3
1
Dept. of Industrial engineering and Management. Sri Siddhartha Institute of Technology,
Tumkur, India
2
Department of Mechanical engineering. UBDT. College of Engineering, Davangere
India
3
Dept. of Mechanical engineering, Sri Siddhartha Institute of Technology, Tumkur
Corresponding author: [e-mail]
ABSTRACT
Taguchi methods have proved to be successful over the last two decades for improvement of
product quality and process performance. This study is carried out to simultaneously
optimize the tribological properties: wear rate and frictional force of aluminum metal matrix
composite. Al-Cu-Mg alloy reinforced with 6 Wt % of titanium dioxide was prepared using
stir casting method. Dry sliding wear test was conducted to understand the tribological
behavior of samples. The experiments were conducted as per the Taguchi design of
experiment. The wear parameters chosen for the experiment were: sliding speed and load
and sliding distance. Each parameter was assigned three levels. The experiment consists of
27 tests according to L
27
orthogonal array. Signal to noise ratio analysis has been carried
out to determine optimal parametric condition, which yields minimum wear rate and
frictional force. Harrington’s desirability functional method is adopted for multifunctional
optimization of tribological parameters and the confirmation experiments were conducted to
verify the predicted model.
Key words: Metal matrix composite; Titanium Dioxide; Taguchi Technique; Signal to noise
ratio; significant factors.
1. INTRODUCTION
Particulate reinforced metal matrix composites (PR-MMCs) have combination of low
density, improved stiffness & strength, high wear resistance and isotropic properties [1-2]
Aluminum metal matrix composites (AMCs) reinforced with hard ceramic particles has
emerged as a potential material for wear – resistance and weight critical applications, such as
brake drums, cylinder liners, pistons, cylinder blocks, connecting rods and so on [3]. These
new materials offer promising perspectives in assisting automotive engineers to achieve
improvement in vehicle fuel efficiency. Several engineering applications require enhanced
friction and wear performances. The principle tribological parameter that affects the friction
and wear performance of discontinuously reinforced aluminum composites are surface
interaction, mechanical characteristic (extrinsic to materials), material characteristic (intrinsic

1180 Hemanth Kumar.T.R. et al Vol.10, No.12
to materials) and tribo contact condition. Sannino et al., discussed most frequently
encountered factors which are associated with four tribological parameters in his review
paper [4] and these factors are shown in the Cause and Effect diagram in Figure 1.
Mechanical
Parameters Tribocontact
Surface InteractionMaterial Parameters
Wear & Friction
Mechanism
Loading Conditions
Relative Velocity
Intial
Temperature
Surface FinishDuration of interaction
Contact Geometry
Topography
evolution
Third Body
Lubrication
Dry
Chemical Reaction
Solid Film
Transfer
Flash temp. rising
Rolling
Sliding
Fretting
Reciprocating
Impact
Erosion
Types of
chemical bonding
Melting point/ Phase transfer
Microstructure
Precipitation hardening
Composite
/Second phase
Hardness / Yield
strength
Fracture
Toughnes
Fracture
toughness
Thermal
poperties
Figure1. Cause and effect diagram of tribological properties.
Some of the researcher investigated the effect of particulate reinforcement on aluminum alloy
and reported that composites exhibits higher wear resistance, higher seizure pressure, less
frictional heating and marginally lower coefficient of friction when compared with the matrix
alloy [5-7]. At elevated temperature, Al7075 – glass fiber reinforced composites exhibits
better wear resistance than the base alloy [8]. Sliding wear, slurry erosive wear of the as-cast
metal matrix composites (MMCs) were studied by Ramachandra et al., [9] and reported that
sliding wear and slurry erosive wear resistance improved considerably with the addition of
SiC particles, where as corrosion resistance decrease.
Dry sliding wear behavior of aluminum metal matrix composites has been reported [10-12]
and abrasive wear of aluminum composites has extensively reviewed by Deuis et al., [13].
Uyyuru et al [14] studied the effect of reinforcement volume fraction and size distribution on
the tribological behavior of Al-composites and Martin investigated the temperature effects on
the wear behavior of particulate reinforced Al-based composites [15] and influence of heat
treatment on the wear behavior of aluminum composites by Gomez de Salazar et al [16].
Effect of aging on wear of MMCs was carried out by Guo et al [17] and Grigoris et al [18].
Much research has been carried out to understand wear behavior of composite materials.
Meager information is available regarding the simultaneous optimization of tribological
parameters: Sliding wear rate and frictional force of the Metal matrix composite. In this light,
this study is carried out to optimize tribological parameters of Titanium Dioxide reinforced
Aluminum metal matrix composites using Taguchi’s parameter design methodology.
1.1 Taguchi Design of Experiment
Taguchi design of experiment is a powerful analysis tool for modeling and analyzing the
influence of control factors on performance output. The most important stage in the design of

Vol.10, No.12 Taguchi Technique 1181
experiment lies in the selection of the control factors. Therefore, a number of factors are
included so that non-significant variables can be identified at earliest opportunity. Taguchi
method provides a simple, systematic and efficient methodology for the optimization of the
control factors.
Quality characteristic of a product under investigation in response to a factor introduced in
the experimental design is the ‘Signal’ of the desired effect. The effect of the external factors
(Uncontrollable factors) on the outcome of the quality characteristic under test is termed as
‘noise’. The Signal-to-Noise ratio (S/N ratio) measures the sensitivity of the Quality
characteristic being investigated in a controlled manner to those of external influencing
factors (Noise factors) not under control. The S/N ratio is a transformed figure of merit,
created from the loss function. S/N ratio combines both the parameters (the mean level of the
quality and the variation around this mean) in a single metric. The aim in any experiment is
always to determine the highest possible S/N ratio for the result (wear rate) a high value of
S/N ratio implies that signal is much higher than the random effects of noise factors [19].
2. EXPERIMENTAL DETAILS.
2.1 Test Materials
2.1.1. Matrix material.
The aluminum alloy AA2618 is used as a matrix material for the preparation of composite
material. This alloy is a heat treatable Al-Cu-Mg-Fe-Ni alloy, developed for high temperature
applications [20], especially in the manufacture of aircraft engine components and
automobile applications [21]. This alloy has good elevated temperature strength up to 204
O
C.
The presence of small amounts of Fe and Ni produces micro structural stability under thermal
exposure [20]. This alloy derives its strength from a combination of precipitation and
dispersion hardening. The composition of this alloy is Cu - 2.18 %, Mg - 1.43%, Ni -1.1%, Fe
- 0.93%, Si - 0.16%, Ti - 0.04 %, Mn - 0.028% and balance is aluminum.
2.1.2. Reinforcement material.
Titanium dioxide (TiO
2
) is used as a reinforcing material for the preparation of composite
material. Titanium dioxide is a fine white powder, which is one of the most important
reinforcing materials used for making composites for many engineering applications. Against
this background, the present research work has been undertaken, with an objective to
optimize the tribological behavior of TiO2 reinforced in Aluminum alloy composites. The
composites were produced using liquid vortex method by adding 6 wt % of Titanium dioxide
in aluminum alloy matrix material.
2.2 Plan of experiments using orthogonal array.
Tribological behaviors of the samples were studied by conducting the dry sliding wear test as
per the standard orthogonal array (OA). The wear parameters chosen for the experiment
were: sliding speed in m/s, load in N and sliding distance in m. The non-linear behavior of
the process parameters, if exists, can only be revealed if more than two levels of the
parameters are investigated. Therefore, each parameter was analyzed at three levels. The
process parameters along with their values at three levels are given in Table 1. It was also
decided to study the two factor interaction effects on tribological behavior of the sample. The

1182 Hemanth Kumar.T.R. et al Vol.10, No.12
selected interactions were: (i) Sliding speed and load (ii) sliding speed and distance and (iii)
Load and distance. An L
27
(3
13
) OA having 26 degrees of freedom (DOF) was selected for the
conduction of experiment. The first column in the OA was assigned to Sliding speed, second
column was assigned to load and the fifth column was assigned to sliding distance. The
remaining columns were assigned to their interactions.
Table1. Process parameters with their different levels.
2.3. Experimental Procedure
To evaluate the tribological performance of the composites under dry sliding condition, wear
tests were carried out on a pin-on-disc type friction and wear monitoring test rig as per
ASTM G 99 standard. The experimental set up is shown in Figure2. The counter body is a
disc made of hardened ground steel (EN-32, hardness 64 HRC, surface roughness 0.5 Ra).
The specimen is held stationary and the disc is rotated while a normal force is applied
through a lever mechanism. The wear test was conducted as per the orthogonal array of
Taguchi as shown in Table 2. The wear rate of the specimen was studied as a function of the
sliding velocity, applied load and sliding distance. Wear test were conducted as per
procedure reported in the paper [12] and the wear rate and the frictional force are noted. At
the end of each experiment, the specimen was cleaned with acetone and dried, so that disc is
free from wear debris. Each experiment was repeated twice and mean response values are
tabulated in Table 2.
pulley
LVDT
specimen
Disc
Base plate
Weights
Damper
Motor Frame
Figure 2 Pin on disc test rig.
Factors Code Units Level 1 Level 2 Level 3
Sliding speed S m/s 1.256 2.090 3.056
Load L N 19.6 29.4 39.2
Sliding Distance D m 600 1200 1800

Vol.10, No.12 Taguchi Technique 1183
Table 2 S/N ratio for Dry Sliding Wear & Frictional Force.
Test Sliding speed
(m/s)
Load
(N)
Sliding
Distance
(m)
Wear rate
(mm
3
/N-m)
Frictional
Force
( N)
S/N
Ratio
(db)
wear rate
S/N Ratio
(db)
Frictional
force
1 1.256 19.6 600 24.0 8.50 -27.6042 -18.5884
2 1.256 19.6 1200 57.5 8.15 -35.1934 -18.2232
3 1.256 19.6 1800 63.0 8.90 -35.9868 -18.9878
4 1.256 29.4 600 27.0 11.60 -28.6273 -21.2892
5 1.256 29.4 1200 57.0 12.45 -35.1175 -21.9034
6 1.256 29.4 1800 82.0 12.40 -38.2763 -21.8684
7 1.256 39.2 600 78.0 12.40 -37.8419 -21.8684
8 1.256 39.2 1200 82.0 14.50 -38.2763 -23.2274
9 1.256 39.2 1800 91.0 17.40 -39.1808 -24.8110
10 2.090 19.6 600 19.0 8.30 -25.5751 -18.3816
11 2.090 19.6 1200 56.5 7.50 -35.0410 -17.5012
12 2.090 19.6 1800 75.0 5.80 -37.5012 -15.2686
13 2.090 29.4 600 27.0 10.60 -28.6273 -20.5061
14 2.090 29.4 1200 62.0 9.80 -35.8478 -19.8245
15 2.090 29.4 1800 83.0 10.80 -38.3816 -20.6685
16 2.090 39.2 600 42.0 12.90 -32.4650 -22.2118
17 2.090 39.2 1200 65.0 14.30 -36.2583 -23.1067
18 2.090 39.2 1800 84.0 14.20 -38.4856 -23.0458
19 3.056 19.6 600 23.0 5.30 -27.2346 -14.4855
20 3.056 19.6 1200 47.0 7.50 -33.4420 -17.5012
21 3.056 19.6 1800 53.0 7.10 -34.4855 -17.0252
22 3.056 29.4 600 36.0 10.90 -31.1261 -20.7485
23 3.056 29.4 1200 45.0 10.50 -33.0643 -20.4238
24 3.056 29.4 1800 53.0 10.65 -34.4855 -20.5470
25 3.056 39.2 600 34.5 14.80 -30.7564 -23.4052
26 3.056 39.2 1200 66.0 14.05 -36.3909 -22.9535
27 3.056 39.2 1800 72.5 13.90 -37.2068 -22.8603
3. RESULT DISCUSSION
3.1 Signal-to-Noise Ratio
The experimental observations are transformed into Signal-to-Noise ratio. There are several
S/N ratios available depending on the type of characteristics under study. The wear rate and
frictional force are coming under ‘smaller is better’ type of quality characteristic and the
respective S/N ratio is calculated using formula1.
----------- (1)
Where n = number of tests in a trial.
For the present study n = 2.
The aim of this experiment is to determine the highest possible Signal-to-noise ratio for the
parameters under study. A high value of S/N ratio implies that signal is much higher than the

1184 Hemanth Kumar.T.R. et al Vol.10, No.12
random effects of noise factors. The S/N ratio was computed using equation (1) for each of
the 27 trial and the values are listed in Table 2.
3.2 Analysis of Variance
Analysis of variance (ANOVA) is used to analyze the influence of wear parameters like
Sliding speed, sliding distance and applied load on the tribological performance
characteristics: wear and frictional force. This analysis was carried out for the level of
Significance of 5 % with 95 % confidential level. Table 3 and Table 4 shows the results of
ANOVA analysis for Sliding wear and Frictional force of the composite material
respectively. The total sum of squares value is used to measure the relative influence of the
factors. The large value of sum of squares, the more influential the factor is for controlling
the responses. These values are used to determine the percentage contribution factors. From
the table 4 it is found that distance traveled ( P= 57.02 % ) is the most significant factor,
whereas load as 18.69 % contribution and sliding speed has 7.629 % contribution towards
the sliding wear quality characteristic under study. However, the interaction between Sliding
speed X load is 3.95%, speed X distance is 2.90 % and load X distance has 1.0 %
contribution. Total error associated in this analysis is approximately about 8.78 %. Table5
shows the analysis of variance for Frictional force of the composite. It is found that load
applied on the test material is the most significant factor which induces the frictional force,
whereas sliding speed as 3.69 % contribution and distance as negligible contribution towards
frictional force induced by the test specimen. However, the interaction between Sliding speed
X distance is 1.43 % whereas, sliding speed X load and load X distance has negligible
contribution to frictional force. Total error associated in this analysis is approximately about
8.15 %.
Table3. ANOVA test for Sliding Wear rate of TiO2 reinforced composite material.
Factors DOF Seq SS Adj SS Adj MS F P
%
Contribut
ion
Sliding speed
(S)
m/s
2 984.02 984.02 492.01 6.58 0.020 7.629
Load(L) N 2 2302.74 2302.74 1151.37 15.40 0.002 18.69
Distance(D) m 2 6870.91 6870.91 3435.45 45.94 0.000 57.02
S*L 4 545.59 545.59 136.40 1.82 0.218 3.95
S*D 4 421.26 421.26 105.31 1.41 0.315 2.90
L*D 4 194.37 194.37 48.59 0.65 0.643 1.00
Error 8 598.30 598.30 74.79
8.78
Total 26 11917.19
11917.19
3.3. Simultaneous Optimization of Tribological Properties.
Many researchers used Taguchi’s parameter design for optimizing a single quality
characteristic. During the optimization of multiple quality characteristics, the objective is to

Vol.10, No.12 Taguchi Technique 1185
determine the best factors settings which will simultaneously optimize all the quality
characteristics under study. This investigation was carried out to simultaneously optimize the
tribological parameters; sliding wear rate and frictional force. The experimental results are
analyzed to determine the process parameters which yield the least sliding wear and
minimum frictional force.
Table4. ANOVA test for Frictional Force for TiO2 reinforced composite material.
Factors DOF SS Adj SS V F P
%
Contribut
ion
Sliding speed
(S)
m/s
2 10.416 10.416 10.416 3.67 0.074 3.69
Load(L) N 2 209.724 209.724 209.724 73.87 0.000 85.47
Distance(D) m 2 1.922 1.922 1.922 0.68 0.535 0.205
S*L 4 1.884 1.884 1.884 0.33 0.849 0.190
S*D 4 4.914 4.914 4.914 0.87 0.524 1.43
L*D 4 3.484 3.484 3.484 0.61 0.665 0.846
Error 8 11.356 11.356 11.356 8.155
Total 26 243.70 243.70
The least sliding wear rate is recorded, during 10
th
test run, when the process parameters were
at 2
nd
level of Sliding speed and first level of load and sliding distance. The second quality
characteristic under study; frictional force is minimum, when the process parameters are at
3
rd
level of sliding speed and first level of load and sliding distance during the 19
th
test run.
Hence, the optimum combinations of the factors levels have not been identical for the two
performance characteristics. Achieving both the optimum criteria simultaneously is
impossible. Hence, a suitable trade off between the two becomes inevitable. Here,
Harrington’s desirability function method [22, 23] has been adopted for multi response
optimization. The response variable Y
i
can be transformed to a desirability value h
i
with the
help of the desirability function. Harrington’s had developed a functional approach using
desirability function to optimize the multi response situation. The one sided transformation as
proposed by Harrington can be represented as
h
i
= exp [-exp (-Y
i )
] -------(2)
Individual desirability of all the responses can be combined to get a Single value of
desirability by the expression
……. (3)
Table 5 shows the optimum levels of responses. For Harrington desirability function, the
composite desirability H for two responses has been computed for both the settings and then
large of these has been identified as the optimum operating combination of the levels. The
composite desirability for the settings has been obtained as shown in the Table6, from which
it has been obvious that the setting S2, L1 and D1 has been found to be the optimal setting for
optimizing the tribological properties.

1186 Hemanth Kumar.T.R. et al Vol.10, No.12
Table 5. Concurrent optimization of multiple factors
Optimal levels
S L D
Sliding wear
rate
Individual
Desirability
(h
1
)
Frictional
force
Individual
Desirability
(h
2
)
Combined
Desirability
(H)
2.09 19.6 600 19.0 0.999 8.3 0.999 0.999
3.056 19.6 600 23.0 1 5.3 0.995 0.997
3.4 Confirmation Experiment
The confirmation experiment is conducted to verify that, the optimal setting factors derived
previously will actually yield an improvement in quality characteristic under study and how
close are the respective predictions with the real ones. However, if the observed S/N ratios
under the optimum conditions differ drastically from their respective predictions, the additive
model fails to be a failure eventually.
Table 6 shows the comparison of the predicted wear rate and frictional force with the actual
response of the quality characteristic under study. Deviation finds to be less from the
predicted values. The error is calculated as the difference between actual and predicted values
of S/N ratio.
Table 6 Verification of test result
4. CONCLUSIONS
• Dry sliding wear and frictional force of the composite material under different loads and
Sliding velocities can be successfully analyzed using Taguchi design of experiment
• The analysis of variance for Sliding wear rate of the composite material shows that, that
distance traveled by the specimen is the most significant factor, whereas load and
sliding speed has little contribution towards the Quality characteristic under study.
• The analysis of variance for Frictional force of the composite material shows that, load
is the most significant factor which induces the frictional force, whereas sliding speed
and distance has least contribution towards frictional force induced by the test
specimen.
Sliding Wear rate Frictional force
Predicted
optimum
Actual
optimum Predicted optimum Actual optimum
Parameters
Level
S: 3.056,
l: 19.6, D: 600
S: 3.056,
l: 19.6, D: 600
Parameters
Level
S: 3.056,
l: 19.6, D: 600
S: 3.056,
l: 19.6, D: 600
Wear rate 23.0 22.0 Frictional
force 5.3 5.4
S/N ratio -27.2346 -26.8485 S/N ratio -14.4855 -14.6475
Predicted error of S/N ratio
(db) 0.386103 Predicted error 0.16235
Confidential limit (2σ) 0.546 Confidential limit (2σ) 0.427

Vol.10, No.12 Taguchi Technique 1187
• Taguchi Design of experiment is extended further for the simultaneous optimization of
wear parameters using Harrington’s desirability method. Using Harrington’s desirability
method optimum condition is found and confirmation experiment was conducted to show
that additive models are correct. This method can be adopted for the simultaneous
optimization of Tribological parameters, where both the quality characteristic is important
for the end– product which increases the performance of the product.
REFERENCES
1. Mingzhao Tan, Qibin Xin, Zhenghua Li, B. Zong, Y. Influence of SiC and Al2O3
particulate reinforcements and heat treatments on mechanical properties and damage
evolution of Al-2618 metal matrix composites. Journal of Materials science 36 (2001)
2045 -2053.
2. Nitsham. .A. E. New application for Aluminum based metal matrix composites.
1997. Vol. 54. Light metal age. USA.
3. Debdas Roy. Bikramjit Basu. Amitava Basu Mallick. Tribological properties of Ti-
aluminide reinforced Al-based In- Situ –metal matrix composite. Intermetallics 13
(2005) 733 – 740.
4. A. P. Sannino, H.J. Rack.. Dry sliding wear of discontinuously reinforced aluminum
composites: review and discussion. Wear 189 (1995) 1-19.
5. R.N. Rao. S.Das and P.V. Krishna. Experimental investigation on the influence of SiC
particulate reinforcement in Al alloy composites.
6. Prasanna Kumar.M. Sadashivappa.K. Prabhukumar.G.P. and Basavarajappa.S. Dry
sliding wear behavior of Garnet reinforced metal matrix composite. Materials science.
Vol.12. No.3. 2006. PP. 209 -213.
7. G.Ranganath. S.C. Sharma and M. Krishna. Dry sliding wear of Garnet reinforced
Zink / Al MMCs. Wear. 251 (2001) 1408 – 1413.
8. S.C. Sharma. M.Krishna. P.V.Vizhian and A.Shashishankar. Thermal effects on Mild
wear behavior of al 7075-glass fiber reinforced composite. Vol.216. No.12/2002. PP.
975-982.
9. Ramachandra M, Radhakrisna. Sliding wear, slurry erosive wear and corrosive wear
of Al /SiC composites.Material science. Vol. 24. No. 2/1.2006.
10. G.Ranganath. S.C. Sharma and M. Krishna. Dry sliding wear of Garnet reinforced
Zink / Al MMCs. Wear. 251 (2001) 1408 – 1413
11. Wilson S, alpas. A.T. Wear mechanism map for metal matrix composites. Wear.
1997; 212: 41-49.
12. S. Basavarajappa. G. Chandramohan. R. Subramanian. Chandrasekar. Dry sliding
wear behavior of Al 2219 / SiC metal matrix composites. Materials science Poland.
Vol. 24. No. 2 /1. 2006.
13. R.L. Deuis. C. Subrananian and J. M. Yellup. Abrasive wear of Aluminum
composites. A review. Wear. 201. (1996) 132 – 144.
14. R.K. Uyyuru, M.K. Surappa. S. Brusethaug. Effect of reinforcement Vol fraction and
size distribution on the tribological behavior of Al - composite. Wear. 260 (2006)
1248 – 1255.
15. A Martin, J Rodriguez, J. Lorea. Temperature effects on the wear behavior of
particulate reinforced Al-based composites. Wear. 255-229 (1999) 615-620.
16. J.M Gomez de, Salazar, M.I. Barrena. Influence of heat treatments on the wear
behavior of an AA 6092 / SiC composites. Wear. 256 (2004) 286 -293.

1188 Hemanth Kumar.T.R. et al Vol.10, No.12
17. J Guo, X Yuan. The aging behavior of SiC /Gr/6013 Al composites in T-4 –T-6
treatments.
18. Grigoris E. Kiourtsidis, stefanos M. Skolianos. Wear behavior of artificially aged
AA2024 /SiC composites in comparison with conventionally wear resistance ferrous
materials. Wear. 253 (2002) 946 -956.
19. Phillip. J. Ross. Taguchi techniques for quality engineering. Mc. Graw hill, New
York. 1-179. (2005)
20. J.H. Hatch (ed), Aluminum: properties and physical metallurgy, ASM, Metals park,
OH, (1988), 371.
21. James C. Williams. Progress in structural materials for Aerospace systems. Acta
materials. 51 (2003) 5775 – 5799.
22. Gaitonde v.n, Achyutha.B.T, Siddeshwarappa.B. Burr size minimization in drilling
using taguchi technique. Indian Journal of engineering & Materials. Vol. 12. April
2005. PP. 91-96.
23. Derringer G & Suich R. Simultaneous optimization of several response variables.
Journal of quality technology. 12 (1980) 214 -219.